System and method for processing distorted or inaccurate input data obtained from an environment

Abstract

A regression planning and evaluation system that, when operated, obtains audit evidence, for example from manufacturing facility. The system comprises a server arrangement configured to (a) obtain input data, input data comprising data values of variables for multiple samples; (b) validate input data; (c) provide user with user interface to input audit parameters, audit parameters comprising at least one of: assurance level, tolerable error, statistical sampling technique; (d) generate regression plot and identify upper and lower acceptance bounds indicated with respect to regression line, regression plot represents variation of target variable with explanatory variable for samples and (e) identify key items to be tested, key items being samples falling outside the aforesaid acceptance bounds. Audit evidence, for example, can initiate further input data collection, further testing and so forth for reducing errors or distortions present in input data that was earlier acquired for analysis by system.

What is claimed is: 
     
       1. A regression planning and evaluation system that processes input data obtained from an environment, wherein the regression planning and evaluation system, when operated, obtains audit evidence by detecting distortions or inaccuracies of the input data, characterized in that the regression planning and evaluation system comprises a server arrangement that is configured to:
 (a) obtain input data pertaining to a given use case for which the audit evidence is to be obtained, the input data comprising data values of a set of variables for a plurality of samples associated with the given use case; 
 (b) validate the input data based on a predefined set of statistical rules; 
 (c) provide a user with an interactive user interface to enable the user to input a plurality of audit parameters to be employed for the audit evidence, the plurality of audit parameters comprising at least one of: an assurance level to be achieved, a tolerable error, a statistical sampling technique to be employed; 
 (d) generate a regression plot to be presented to the user via the interactive user interface and identify an upper acceptance bound and a lower acceptance bound to be indicated with respect to a regression line in the regression plot, the regression plot representing a variation of a target variable with respect to at least one explanatory variable, wherein the regression plot is to be generated for the plurality of samples, and wherein the upper acceptance bound and the lower acceptance bound are to be identified based on the plurality of audit parameters; and 
 (e) identify, from amongst the plurality of samples, key items that are required to be tested, the key items being samples that fall outside the upper acceptance bound and the lower acceptance bound in the regression plot, 
 
       wherein the server arrangement is further configured to provide an optimisation option in the interactive user interface that, when selected by the user, optimises the plurality of audit parameters in a manner that the number of key items that are required to be tested is reduced. 
     
     
       2. The regression planning and evaluation system of  claim 1 , wherein an identification of the key parameters performed by the system causes the system to acquire further input data to resolve distortions or inaccuracies associated with the key items, and wherein the further data is obtained from at least one of: manual data entry, sensor signals, equipment measurements, image capture, database access. 
     
     
       3. The regression planning and evaluation system of  claim 1 , wherein the server arrangement is coupled to a regression module coupled in communication with the server arrangement for performing regression computations. 
     
     
       4. The regression planning and evaluation system of  claim 3 , wherein the regression module is implemented by employing a configuration of field-programmable gate arrays (FPGA's) and reduced instruction set (RISC) computers. 
     
     
       5. The regression planning and evaluation system of  claim 4 , wherein the reduced instruction set (RISC) computers are reconfigured in response to a structure of data received thereat to be processed, allocating one or more regression computations to a given reduced instruction set (RISC) computer, and by performing correlations by way of data exchange between a plurality of the reduced instruction set (RISC) computers. 
     
     
       6. The regression planning and evaluation system of  claim 1 , wherein the server arrangement is configured to:
 enable, via the interactive user interface, the user to adjust the assurance level to be achieved and/or the tolerable error for the audit evidence; and 
 iteratively perform (d) and (e) for the adjusted assurance level and/or the adjusted tolerance error. 
 
     
     
       7. The regression planning and evaluation system of  claim 1 , wherein the server arrangement is configured to enable, via the interactive user interface, the user to reconcile the validated input data to a source from where the input data was obtained at (a). 
     
     
       8. The regression planning and evaluation system of  claim 1 , wherein the server arrangement is configured to determine relationships existing between the variables of the set. 
     
     
       9. The regression planning and evaluation system of  claim 8 , wherein the server arrangement is configured to:
 enable, via the interactive user interface, the user to select a variable of interest from the set of variables; 
 determine an impact of remaining variables of the set on the variable of interest, based on the relationships existing between the variables; and 
 represent, via the interactive user interface, the impact of the remaining variables on the variable of interest. 
 
     
     
       10. The regression planning and evaluation system of  claim 9 , wherein the server arrangement is configured to represent, via the interactive user interface, inter-relations that exist between the remaining variables. 
     
     
       11. The regression planning and evaluation system of  claim 8 , wherein the server arrangement is configured to:
 enable, via the interactive user interface, the user to select, from amongst the set of variables, the target variable and a plurality of explanatory variables; 
 determine an impact of the plurality of explanatory variables on the target variable, based on the relationships existing between the variables; and 
 represent, via the interactive user interface, a residual plot indicative of the impact of the plurality of explanatory variables on the target variable. 
 
     
     
       12. A method for obtaining audit evidence, the method comprising using regression planning and evaluation system that processes input data obtained from an environment, wherein the regression planning and evaluation system, when operated, obtains audit evidence by detecting distortions or inaccuracies of the input data, wherein the method includes:
 (a) obtaining input data pertaining to a given use case for which the audit evidence is to be obtained, the input data comprising data values of a set of variables for a plurality of samples associated with the given use case; 
 (b) validating the input data based on a predefined set of statistical rules; 
 (c) providing a user with an interactive user interface to enable the user to input a plurality of audit parameters to be employed for the audit evidence, the plurality of audit parameters comprising at least one of: an assurance level to be achieved, a tolerable error, a statistical sampling technique to be employed; 
 (d) generating a regression plot to be presented to the user via the interactive user interface and identifying an upper acceptance bound and a lower acceptance bound to be indicated with respect to a regression line in the regression plot, wherein the regression plot represents a variation of a target variable with respect to at least one explanatory variable, the regression plot being generated for the plurality of samples, the upper acceptance bound and the lower acceptance bound being identified based on the plurality of audit parameters; and 
 (c) identifying, from amongst the plurality of samples, key items that are required to be tested, the key items being samples that fall outside the upper acceptance bound and the lower acceptance bound in the regression plot, 
 
       wherein the method further comprises providing an optimisation option in the interactive user interface that, when selected by the user, optimises the plurality of audit parameters in a manner that the number of key items that are required to be tested is reduced. 
     
     
       13. The method of  claim 12 , wherein the method further includes arranging for the server arrangement to be coupled to a regression module that is coupled in communication with the server arrangement for performing regression computations. 
     
     
       14. The method of  claim 13 , wherein the method further comprises arranging for the regression module to be implemented by employing a configuration of field-programmable gate arrays (FPGA's) and reduced instruction set (RISC) computers. 
     
     
       15. The method of  claim 14 , wherein the method includes implementing the reduced instruction set (RISC) computers to be reconfigurable in response to a structure of data received thereat to be processed, allocating one or more regression computations to a given reduced instruction set (RISC) computer, and by performing correlations by way of data exchange between a plurality of the reduced instruction set (RISC) computers. 
     
     
       16. The method of  claim 13 , wherein the method further comprises using the regression planning and evaluation system for:
 enabling, via the interactive user interface, the user to adjust the assurance level to be achieved and/or the tolerable error for the audit evidence; and 
 iteratively performing the steps (d) and (e) for the adjusted assurance level and/or the adjusted tolerance error. 
 
     
     
       17. The method of  claim 12 , further comprising using the regression planning and evaluation system for enabling, via the interactive user interface, the user to reconcile the validated input data to a source from where the input data was obtained at the step (a). 
     
     
       18. The method of  claim 12 , wherein the method further comprises using the regression planning and evaluation system for determining relationships existing between the variables of the set. 
     
     
       19. The method of  claim 18 , wherein the method further comprises using the regression planning and evaluation system for:
 enabling, via the interactive user interface, the user to select a variable of interest from the set of variables; 
 determining an impact of remaining variables of the set on the variable of interest, based on the relationships existing between the variables; and 
 representing, via the interactive user interface, the impact of the remaining variables on the variable of interest. 
 
     
     
       20. The method of  claim 19 , wherein the method further comprises using the regression planning and evaluation system for representing, via the interactive user interface, inter-relations that exist between the remaining variables. 
     
     
       21. The method of  claim 18 , wherein the method further comprises using the regression planning and evaluation system for:
 enabling, via the interactive user interface, the user to select, from amongst the set of variables, the target variable and a plurality of explanatory variables; 
 determining an impact of the plurality of explanatory variables on the target variable, based on the relationships existing between the variables; and 
 representing, via the interactive user interface, a residual plot indicative of the impact of the plurality of explanatory variables on the target variable. 
 
     
     
       22. A software product comprising a non-transitory machine-readable data storage medium having stored thereon program instructions that, when executed by a processing device, cause the processing device to:
 (a) obtain input data pertaining to a given use case for which an audit evidence is to be obtained, the input data comprising data values of a set of variables for a plurality of samples associated with the given use case; 
 (b) validate the input data based on a predefined set of statistical rules; 
 (c) provide a user with an interactive user interface to enable the user to input a plurality of audit parameters to be employed for the audit evidence, the plurality of audit parameters comprising at least one of: an assurance level to be achieved, a tolerable error, a statistical sampling technique to be employed; 
 (d) generate a regression plot to be presented to the user via the interactive user interface and identify an upper acceptance bound and a lower acceptance bound to be indicated with respect to a regression line in the regression plot, wherein the regression plot represents a variation of a target variable with respect to at least one explanatory variable, and wherein the regression plot is to be generated for the plurality of samples, further wherein the upper acceptance bound and the lower acceptance bound are to be identified based on the plurality of audit parameters; and 
 (e) identify, from amongst the plurality of samples, key items that are required to be tested, the key items being samples that fall outside the upper acceptance bound and the lower acceptance bound in the regression plot, 
 
       wherein the processing device is further configured to provide an optimisation option in the interactive user interface that, when selected by the user, optimises the plurality of audit parameters in a manner that the number of key items that are required to be tested is reduced. 
     
     
       23. The software product of  claim 22 , wherein the program instructions, when executed by the processing device, cause the processing device to:
 enable, via the interactive user interface, the user to adjust the assurance level to be achieved and/or the tolerable error for the audit evidence; and 
 iteratively perform (d) and (e) for the adjusted assurance level and/or the adjusted tolerance error.